MOLECULAR AND CELLULAR BIOLOGY, Apr. 2008, p. 2271–2282
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
Vol. 28, No. 7
Phosphorylation of Human Jak3 at Tyrosines 904 and 939 Positively
Regulates Its Activity?
Hanyin Cheng,1,2Jeremy A. Ross,2Jeffrey A. Frost,1and Robert A. Kirken2*
Department of Integrative Biology and Pharmacology, University of Texas Health Science Center at Houston Medical School,
Houston, Texas 77030,1and Department of Biological Sciences, University of Texas at El Paso, El Paso, Texas 799682
Received 28 September 2007/Returned for modification 8 November 2007/Accepted 22 January 2008
Janus tyrosine kinase 3 (Jak3) is essential for signaling by interleukin-2 (IL-2) family cytokines and proper
immune function. Dysfunctional regulation of Jak3 may result in certain disease states. However, the molec-
ular mechanisms governing Jak3 activation are not fully understood. In this study, we used a functional-
proteomics approach to identify two novel tyrosine phosphorylation sites within Jak3, Y904 and Y939, which
are conserved among Jak family proteins. By using phosphospecific antibodies, both residues were observed to
be rapidly induced by stimulation of cells with IL-2 or other ?c cytokines. Mechanistic studies indicated that
Y904 and Y939 regulate Jak3 activities. A phenylalanine substitution at either site greatly reduced Jak3 kinase
activity in vitro and its ability to phosphorylate signal transducer and activator of transcription 5 (Stat5) in
vivo, suggesting that phosphorylation of these previously unrecognized residues positively regulates Jak3
activity. Y904 and Y939 were required for optimal ATP usage by Jak3, while phosphorylation of Y939
preferentially promoted Stat5 activity in intact cells. Together, these findings demonstrate positive functional
roles for two novel Jak3 phosphoregulatory sites which may be similarly important for other Jak family
members. Identification of these sites also provides new therapeutic opportunities to modulate Jak3 function.
The Janus kinase (Jak) family of cytoplasmic tyrosine ki-
nases associates with a variety of cell surface receptors to
perform essential roles for transducing intracellular signals (9,
15). There are four Jak family members in vertebrates: Jak1,
Jak2, Jak3, and Tyk2. While Jak1, Jak2, and Tyk2 are ubiqui-
tously expressed, Jak3 is predominantly expressed in hemato-
poietic cells (20, 30, 41). Jak3 specifically associates with the
cytokine receptor ? common (?c) chain and can be activated by
interleukin-2 (IL-2) family cytokines such as IL-2, IL-4, IL-7,
and IL-9 (40, 45). Inhibitory mutations in Jak3 or its binding
partner, ?c, can result in severe combined immunodeficiency
(SCID) syndrome in humans and mice, which is clinically man-
ifested by limited numbers of T, natural killer, and functional
B cells (34, 35). Hyperactivation of Jak3 has also been associ-
ated with diseases such as asthma (31) and cancers of the
immune system (44). The restricted expression and function of
Jak3 has made it a promising target for controlling these dis-
eases (6, 33, 39).
The activation of Jak proteins contributes to multiple cellu-
lar processes, including cell growth, proliferation, and differ-
entiation (1). Following receptor engagement by cytokines, the
activation of Jak proteins is believed to occur by auto- or
transphosphorylation of key tyrosine residues located within
their activation loops (12). Stimulation of hematopoietic cells
with IL-2 family growth factors results in the phosphorylation
and enzymatic activation of ?c-associated Jak3 and another
Jak family member, Jak1, which may bind to a cytokine-specific
receptor subunit cooperatively with ?c (19). Activated Jak1
and/or Jak3 then phosphorylate tyrosine residues on the asso-
ciated receptors to produce docking sites for SH2- or PTB-
containing proteins such as signal transducer and activator of
transcription 5 (Stat5) (14, 24, 25), leading to their phosphor-
ylation and subsequent activation. These proteins then regu-
late many downstream events, including gene transcription.
Phosphorylation plays a critical role in regulating Jak3 ki-
nase activity. It has been reported that two adjacent tyrosines
located in the Jak3 kinase activation loop are phosphorylated
to positively (Y980) or negatively (Y981) regulate its catalytic
activity (47). Phosphorylation of Jak proteins can also provide
binding sites for other signaling molecules. For example, phos-
phorylation of Jak3 on Y785 has been reported to create a
binding site for the adaptor protein SH2B-?, although the
functional significance of this interaction is unknown (23).
Negative regulatory mechanisms of Jak3 activity include de-
phosphorylation by CD45 and T-cell protein tyrosine phos-
phatase (17, 38). Suppressor of cytokine signaling family pro-
teins form a classical negative feedback loop to attenuate
cytokine signaling that can also act through the Jak/Stat path-
To determine whether other phosphosites exist, we mutated
the three known residues, Y980, Y981, and Y785, and found
no significant change in total tyrosine phosphorylation. Using
mass spectrometry, we identified two additional phosphoty-
rosines in Jak3 at Y904 and Y939. Phosphospecific antibodies
confirmed that phosphorylation of Jak3 on these sites occurred
in response to IL-2 and other IL-2 family cytokines in multiple
cell types, including primary human T cells. Phenylalanine
substitution of these residues inhibited Jak3 tyrosine phos-
phorylation and catalytic activity. Evidence is provided to sug-
gest that Y904 is required for ATP binding while Y939 may be
required for substrate association. More importantly, these
sites are conserved in other Jak family members, suggesting a
universal regulatory role for these tyrosine residues.
* Corresponding author. Mailing address: Department of Biological
Sciences, Biosciences Building, University of Texas at El Paso, 500 W.
University Ave., El Paso, TX 79902. Phone: (915) 747-5844. Fax: (915)
747-5808. E-mail: firstname.lastname@example.org.
?Published ahead of print on 4 February 2008.
MATERIALS AND METHODS
Plasmids and recombinant proteins. The human Stat5a clone, the human ?
common chain (?c) clone, pcDNA3.1(?), and pcDNA3.1/GS were purchased
from Invitrogen. The pRL-TK vector was obtained from Promega. The ?-casein–
luciferase reporter plasmid was generated by cloning a triple repeat of the Stat5
consensus site corresponding to the ?-casein gene promoter (5?-AGATTTCTA
GGAATTCAATCC-3?) into the pGL3-Promoter vector (Promega) by using
SacI and XhoI restriction sites (42). The human Jak3 cDNA was kindly provided
by John J. O’Shea (National Institute of Arthritis and Musculoskeletal and Skin
The Jak3 cDNA was subcloned into pcDNA3.1(?) by using EcoRI and XhoI.
Mutant forms of Jak3 were prepared with the QuikChange site-directed mu-
tagenesis kit (Stratagene) according to the manufacturer’s instructions. The
primers used for Jak3 mutant forms were as follows: for K855A, 5?-CTGGTG
GCCGTGGCACAGCTGCAGCACAG-3?; for Y980, 5?-CGCTTGACAAAGA
CTTCTACGTGGTCCGCGAG-3?; for Y981F, 5?-GACAAAGACTACTTCGT
GGTCCGCGAGCCA-3?; for Y785F, 5?-ATCTCTTCAGACTTTGAGCTCCT
CTCAG-3?; for Y904F, 5?-GCTGGTCATGGAGTTTCTGCCCAGCGGC-3?;
for Y939F, 5?-CAAGGGCATGGAGTTCCTGGGCTCCCGC-3?. The cytoplas-
mic region of human ?c cDNA (nucleotides 863 to 1107) was subcloned into
pGEXKG (Amersham Biosciences) by using EcoRI and XhoI. All subclones and
mutations were verified by DNA sequencing.
For production of glutathione S-transferase (GST)-?c, Escherichia coli BL21/
DE3 was transformed with pGEXKG/?c and grown at 37°C to an optical density
at 600 nm of 0.8. Expression was induced with isopropyl-?-D-thiogalactopyrano-
side (IPTG; 400 ?M, 4 h). Bacteria were pelleted and lysed in GST lysis buffer
(20 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1 mM DTT, 1 mM EDTA, 10 ?g/ml
aprotinin, 10 ?g/ml leupeptin, 10 ?g/ml pepstatin A, 1 mM phenylmethylsulfonyl
fluoride). Insoluble proteins were pelleted by centrifugation (35,000 ? g, 30 min,
4°C). GST-?c was purified by incubating the soluble lysate for 1 h at 4°C with 2
ml of a 50% slurry of preswelled glutathione-agarose (Sigma). Bound GST-?c
was precipitated by centrifugation, and the slurry was washed three times with
lysis buffer. GST-?c was eluted from the beads by incubation with 5 mM free
glutathione and then dialyzed overnight in phosphate-buffered saline plus 10%
glycerol and stored in aliquots at ?80°C.
Antibodies. The Jak3 antibody used for immunoprecipitation and immuno-
blotting was raised against a peptide derived from the unique COOH terminus
of human Jak3 (residues 1104 to 1124) as described previously (27). The anti-
pY(904)Jak3 and anti-pY(939)Jak3 polyclonal antibodies were produced by im-
munizing rabbits with the peptides CLVME(pY)LPSG and CKGME(pY)LGSR
conjugated to keyhole limpet hemocyanin, respectively (Sigma-Genosys). The
antiphosphotyrosine antibody 4G10 (anti-pY), the anti-pY Stat5 monoclonal
antibody, and the anti-Stat5 monoclonal antibody were purchased from Upstate
Biotechnology. The Stat5a polyclonal antibody was raised against a peptide
derived from the C terminus of human Stat5a (residues 775 to 794) as described
previously (46). The anti-GST monoclonal antibody was purchased from Santa
Cell culture and treatment. Human YT and HEK293 cell lines were main-
tained in RPMI 1640 (Cellgro) medium containing 10% fetal bovine serum
(Atlanta Biologicals), 2 mM L-glutamine, and penicillin-streptomycin (50 IU/ml
and 50 mg/ml, respectively) as previously reported (43). The IL-2-dependent
T-cell line Kit225 (kindly provided by J. Johnston, Queens University, United
Kingdom) was maintained in the above medium containing 1 nM recombinant
human IL-2 (rhIL-2) and made quiescent before cytokine stimulation by over-
night incubation in IL-2-free medium. Freshly explanted human T lymphocytes
were purified and maintained in the above medium in the presence of phyto-
hemagglutinin (PHA; 1 ?g/ml; Sigma) for 72 h as previously described (21). T
lymphocytes were subsequently made quiescent by washing and incubation for
24 h in RPMI 1640 medium containing 1% fetal bovine serum prior to exposure
to cytokines. For cytokine treatments, YT cells, quiescent Kit225 cells, or human
T lymphocytes were treated with 100 nM rhIL-2 at 37°C for the times indicated.
Quiescent Kit225 cells were also treated with 100 nM rhIL-9 under the same
conditions. The rhIL-2 was obtained from Hoffman-La Roche, and the rhIL-9
was purchased from PeproTech. Transient transfections of HEK293 cells were
performed with Lipofectamine 2000 (Invitrogen) according to the manufac-
turer’s instructions. Cells were harvested 30 h later for protein analysis.
Solubilization of membrane proteins, immunoprecipitation, and Western blot
analysis. Cell pellets were solubilized in Triton lysis buffer (10 mM Tris-HCl [pH
7.6], 5 mM EDTA [pH 8.0], 50 mM NaCl, 30 mM Na4P2O7, 50 mM NaF, 1 mM
Na3VO4, 1% Triton X-100) containing 1 mM phenylmethylsulfonyl fluoride, 5
?g/ml aprotinin, 2 ?g/ml leupeptin, and 1 ?g/ml pepstatin A and clarified by
centrifugation (16,000 ? g, 10 min, 4°C). For immunoprecipitation reactions,
supernatants were rotated with 2 ?l of Jak3 antibody for 2 h at 4°C. The Jak3
immune complexes were captured by incubation for 30 min at 4°C with protein
A-Sepharose beads (Rockland Immunochemicals). The beads were then washed
three times with cold lysis buffer and eluted by boiling in 1? sodium dodecyl
sulfate (SDS) sample buffer (50 mM Tris-HCl [pH 6.8], 100 mM dithiothreitol,
2% SDS, 0.02% bromophenol blue, 10% glycerol [pH 6.8]). Samples were re-
solved by 7.5% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred
to polyvinylidene difluoride (PVDF) membrane (Amersham Biosciences). West-
ern blot analysis was performed with the indicated primary antibodies either
overnight [anti-pY(904)Jak3 and anti-pY(939)Jak3] or for 1 h (other primary
antibodies) at room temperature. Western blot assays were developed with
horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G (IgG;
heavy plus light chains) or goat anti-rabbit IgG (heavy plus light chains; KPL)
and visualized by using enhanced chemiluminescence and X-ray film. Blots were
also incubated with either IRDye800-conjugated anti-rabbit IgG (Rockland Im-
munochemicals) or Alexa Fluor 680 goat anti-mouse IgG (Molecular Probes)
and visualized with the Odyssey infrared imaging system (LI-COR Biosciences).
Enhanced-chemiluminescence blots were quantified with Scion image software.
Infrared-imaged blots were quantified with LI-COR Odyssey 2.0 software. When
reblotting, PVDF membranes were cleaned in stripping buffer (100 mM ?-mer-
captoethanol, 2% SDS, 62.5 mM Tris-HCl [pH 6.7]) at 55°C for 30 min, blocked,
and then probed with a second primary antibody.
In vitro kinase assays and mass spectrometry analysis. HEK293 cells were
transfected with the appropriate expression vectors for Jak3 and lysed in Triton
lysis buffer. Jak3 proteins were immunoprecipitated with anti-Jak3 antibody and
captured by protein A-Sepharose as described above. The beads were washed
three times with cold lysis buffer and once with 100 mM NaCl–10 mM HEPES-
KOH (pH 7.5) and then resuspended in a kinase buffer (20 mM Tris-HCl [pH
7.5], 5 mM MgCl2, 5 mM MnCl2, 100 ?M ATP) in the absence or presence of 2
?g of GST-?c. Reaction mixtures were incubated at 30°C for the indicated times,
and reactions were terminated by adding SDS sample buffer. When determining
the affinity of immunoprecipitated Jak3 proteins for ATP, assays were performed
at 30°C for 20 min in kinase buffer with increasing concentrations of ATP. This
condition was previously determined to be within the linear range for GST-?c
phosphorylation by immunoprecipitated Jak3 (see Fig. 4). Samples were resolved
by 7.5% SDS-PAGE (see Fig. 2B) or by 12% SDS-PAGE (see Fig. 4 and 7), and
tyrosine phosphorylation levels of Jak3 or GST-?c were assessed by Western
blotting with anti-pY antibodies and quantified by infrared imaging.
For phosphorylation site mapping, immunopurified Jak3 from the in vitro
autokinase assay described above was resolved by 7.5% SDS-PAGE and the
resulting Coomassie blue-stained Jak3 protein was excised and stored at ?80°C
before analysis. Liquid chromatography-tandem mass spectrometry analysis was
performed by the Taplin Biological Mass Spectrometry Facility (Harvard Uni-
versity) or the Biomolecule Analysis Core Facility (University of Texas at El
Paso) by following their standard procedures.
Luciferase assays. Subconfluent HEK293 cells in six-well dishes were trans-
fected with a triple repeat of ?-casein–luciferase reporter plasmid (0.2 ?g),
pRL-TK vector (0.1 ?g), Jak3 plasmids (1 ?g), and Stat5a plasmid (1 ?g). Cells
were harvested 30 h after transfection, and firefly and Renilla luciferase activities
were measured with a dual-luciferase assay reporter system (Promega) according
to the manufacturer’s instructions. Equivalent amounts of cell lysates were also
examined for Jak3 and Stat5a expression by Western blotting.
Peptide pull-down assays. The peptides CKGMEY(939)LGSR and CKGME
pY(939)LGSR were dissolved in coupling buffer (0.1 M NaHCO3[pH 8.3], 0.5
M NaCl) and conjugated to activated CNBr-Sepharose beads overnight at 4°C.
After washing with coupling buffer to remove unbound peptides, the remaining
active groups were blocked with 0.1 M Tris-HCl (pH 8.0). Peptide-conjugated
beads were then incubated with cell lysates from IL-2-treated or untreated YT
cells for 1 h at 4°C. Complexes were washed five times with cold Triton lysis
buffer containing 500 mM NaCl and once with regular Triton lysis buffer. Bound
proteins were then eluted with 1? SDS sample buffer, resolved by 7.5% SDS-
PAGE, and analyzed by Western blotting with anti-pY Stat5, anti-Stat5, and
Jak3 is phosphorylated on previously unrecognized tyrosine
residues. Three tyrosine residues in Jak3, Y980, Y981, and
Y785, were previously identified as phosphoregulatory sites
(23, 47). Mutational analysis was used to reexamine the regu-
latory role of these tyrosines and to determine whether Jak3
2272 CHENG ET AL.MOL. CELL. BIOL.
FIG. 1. IdentificationofnovelphosphorylatedtyrosineresiduesinhumanJak3.(A)Jak3isphosphorylatedonnoveltyrosinesitesinHEK293cells.HEK293
cells were transfected with empty vector or plasmids for WT Jak3 or Jak3 mutant forms Y980F, Y980F/Y981F, Y981F, Y785F, Y980F/Y981F/Y785F
(YYY/FFF), and K855A, as indicated. Cells were harvested at 30 h posttransfection and lysed, and Jak3 proteins were immunoprecipitated (IP) from soluble
lysates with anti-Jak3 antibody. The immunoprecipitates were then Western blotted (immunoblotted [IB]) with pY or anti-Jak3 antibodies as indicated. Shown
of monophosphorylated peptides showing site localization to either Y904 (left panel) or Y939 (right panel), as indicated by asterisks. Methionine oxidation is
denoted (#). (D) Domain architecture of human Jak3 with known and newly identified (asterisks) tyrosine phosphorylation sites. Numbers indicate amino acid
residues of human Jak3.
VOL. 28, 2008TYROSINES 904 AND 939 REGULATE Jak32273
has additional unrecognized regulatory tyrosine phosphoryla-
tion sites. Jak3 Y980, Y981, and Y785 were mutated to phe-
nylalanine individually or in combination and transfected into
HEK293 cells. The transfected Jak3 proteins were then immu-
noprecipitated and immunoblotted with anti-pY or anti-Jak3
antibodies (Fig. 1A). We observed that the Y980F mutant
form of Jak3 and wild-type (WT) Jak3 had comparable levels
of tyrosine phosphorylation (lanes 2 and 3). Similar to earlier
findings, we observed that the Y981F mutant form exhibited
elevated tyrosine phosphorylation (lane 5), as did the Y980F/
Y981F mutant form (lane 4), suggesting that Y981 functions as
a negative regulator in controlling Jak3 activity. We also ob-
served that the tyrosine phosphorylation of the Y785F mutant
form (lane 6) was decreased by 50% compared to WT Jak3,
indicating that this site may account for a substantial portion of
the overall tyrosine phosphorylation of transfected Jak3. Strik-
ingly, the Jak3 Y980F/Y981F/Y785F triple-mutant form was
also readily tyrosine phosphorylated (lane 7), strongly suggest-
ing that novel Jak3 tyrosine phosphorylation sites are present
within this enzyme. Since the Jak3 kinase-inactive K855A mu-
tant form was not tyrosine phosphorylated (lane 8), the phos-
phorylation of WT Jak3 and Y980F/Y981F/Y785F mutant
Jak3 is presumably mediated through autophosphorylation at
distinct Jak3 sites within this model system.
To identify these novel tyrosine phosphorylation sites, WT
Jak3 was immunoprecipitated from transfected HEK293 cells
and subjected to an in vitro autokinase assay for 30 min. The
sample was separated by 7.5% SDS-PAGE and visualized by
Coomassie blue staining. The corresponding 116-kDa band
was excised and subjected to trypsin digestion, and the result-
ing peptides were analyzed by liquid chromatography-tandem
mass spectrometry. Two novel Jak3 phosphorylated peptides
were identified with the Sequest search algorithm. R.LVMEY
LPSGCLR.D contained phosphorylated Y904 (underlined),
and K.GMEYLGSR.R harbored phosphorylated Y939 (un-
derlined) as a second site (Fig. 1B). Tandem mass spectra for
each peptide are shown in Fig. 1C. The positions of these novel
and previously identified Jak3 phosphorylation sites are shown
in Fig. 1D.
Generation of anti-pY(904)Jak3 and anti-pY(939)Jak3
phospho-specific antibodies. To verify that Jak3 is phosphory-
lated at tyrosine 904 and tyrosine 939 and to confirm the regula-
tory roles of these phosphorylation sites, two phospho-specific
antibodies [anti-pY(904)Jak3 and anti-pY(939)Jak3] were gener-
ated. Dot blot analysis was performed with the immunizing phos-
phopeptides and the corresponding nonphosphorylated peptides
(see Materials and Methods for sequences) to determine if the
Jak3 phospho-specific antibodies cross-react with regions distal to
the phosphorylated tyrosines. Increasing amounts of Y904 and
pY904 peptides (Fig. 2A, left panel, top) or Y939 and pY939
peptides (Fig. 2A, left panel, bottom) were spotted onto PVDF
membranes and immunoblotted with the anti-pY(904)Jak3 or
anti-pY(939)Jak3 antibody. The anti-pY(904)Jak3 and anti-
pY(939)Jak3 antibodies recognized the corresponding phos-
phorylated peptides but not the nonphosphorylated counterparts,
indicating that these two phospho-specific antibodies do not
cross-react significantly with the nonphosphorylated peptide. Be-
cause of the amino acid similarities between these two peptides
(60%), their specificity was tested against the corresponding im-
munizing phosphopeptides. Increasing amounts of peptides con-
taining pY904 or pY939 were spotted onto PVDF membranes
and immunoblotted with the anti-pY(904)Jak3 (Fig. 2A, right
panel, top), anti-pY(939)Jak3 (Fig. 2A, right panel, middle), and
anti-pY (Fig. 2A, right panel, bottom) antibodies. The anti-
pY(904)Jak3 antibody recognized the pY904 but not the pY939
peptide, thus validating its specificity. Similarly, the anti-
pY(939)Jak3 antibody recognized the pY939 but not the pY904
the same amount of each peptide was spotted onto the mem-
To further characterize these phospho-specific antibodies,
FIG. 2. Characterization of anti-pY(904)Jak3 and anti-pY(939)Jak3
selective targets. Increasing amounts of Y904 and pY904 peptides (left
spotted onto PVDF membranes and tested for recognition by rabbit
anti-pY(904)Jak3 and anti-pY(939)Jak3 antibodies by Western blotting
(immunoblotting [IB]). Additionally, increasing amounts of pY904 and
pY993 peptides were spotted onto PVDF membranes and tested for
recognition by rabbit anti-pY(904)Jak3 (right panel, top), anti-
pY(939)Jak3 (right panel, middle), and anti-pY (right panel, bottom)
antibodies by Western blotting. (B) Anti-pY(904)Jak3 antibody recog-
WT Jak3 or Y904F or kinase-inactive K855A mutant Jak3. Jak3 proteins
were immunoprecipitated (IP) with anti-Jak3 antibody, followed by an in
vitro kinase assay in the presence of 100 ?M ATP for 30 min as described
in Materials and Methods. Phosphorylation of Y904 was detected by
blotting with anti-pY(904)Jak3 (upper panel). The Jak3 expression level
was monitored by reblotting the membrane with anti-Jak3 antibody
(lower panel). Shown are representative data from three independent
experiments. (C) Anti-pY(939)Jak3 antibody preferentially recognizes ac-
tivated WT Jak3. HEK293 cells were transfected with plasmids for WT or
Y939F or K855A mutant Jak3. Jak3 proteins were immunoprecipitated,
and tyrosine phosphorylation was monitored following an in vitro kinase
assay as described for panel B. The phosphorylation of Jak3 on Y939 was
detected by blotting with anti-pY(939)Jak3 antibody (upper panel). The
blot was reprobed with anti-Jak3 antibody (lower panel). Shown are rep-
resentative data from three independent experiments.
2274CHENG ET AL.MOL. CELL. BIOL.
HEK293 cells were transfected with cDNA encoding WT Jak3
or the Y904F, Y939F, or kinase-inactive K855A mutant form
of Jak3. The Jak3 proteins were then immunoprecipitated and
subjected to an in vitro autophosphorylation reaction. The
Jak3 proteins present in the kinase assays were then resolved
by SDS-PAGE, transferred to PVDF membrane, and exam-
ined for phosphorylation of Y904 and Y939 by Western blot-
ting. As shown in Fig. 2B (lanes 1 and 2), the anti-pY(904)Jak3
antibody recognized autophosphorylated WT Jak3 but not the
Y904F mutant form. Similarly, anti-pY(939)Jak3 preferentially
recognized autophosphorylated WT Jak3 but not the Y939F
mutant form (Fig. 2C, lanes 1 and 2). Reprobing these blots
with anti-Jak3 antibody confirmed similar Jak3 protein expres-
sion (Fig. 2B and C, lower panels). Under the same experi-
mental conditions, neither phospho-specific antibody recog-
nized the kinase-inactive K855A mutant form (Fig. 2B and C,
lanes 3). These data suggest that intact kinase activity is re-
quired for the phosphorylation of Jak3 at Y904 and Y939.
Phosphorylation of Jak3 at Y904 and Y939 occurs in vivo.
?c-containing cytokines such as IL-2, IL-4, IL-7, and IL-9 are
critical for the development and function of the immune sys-
tem (16, 18). Binding of these cytokines to their receptors
results in the rapid phosphorylation of Jak3 and transmission
of their downstream signals to control cellular function (19).
To determine whether Jak3 is phosphorylated at Y904 and
Y939 in response to physiological stimuli, YT cells were stim-
ulated with IL-2 and endogenous Jak3 was then immunopre-
cipitated and examined for phosphorylation (Fig. 3A). We
observed that both tyrosine 904 and tyrosine 939 in Jak3 were
rapidly and transiently phosphorylated following IL-2 stimula-
tion. Phosphorylation of these two tyrosines reached maximal
levels after 2 min and returned to the baseline by 60 min (Fig.
3A). In contrast, the human T-cell leukemia line Kit225
showed that maximum IL-2-induced phosphorylation of Jak3
at Y904 and Y939 occurred at 5 to 10 min and did not return
to basal levels until 120 min post IL-2 stimulation (Fig. 3B). To
FIG. 3. Phosphorylation of Jak3 Y904 and Y939 in YT, Kit225, and primary human T cells is mediated by ?c-containing cytokines. (A) YT cells
were stimulated with IL-2 for the indicated times. Endogenous Jak3 was then immunoprecipitated (IP) with anti-Jak3 antibody and Western
blotted (immunoblotted [IB]) with anti-pY(904)Jak3 (left panels) or anti-pY(939)Jak3 (right panels) antibody. Blots were then reprobed with
anti-pY and anti-Jak3 antibodies. Shown are representative data from three independent experiments. (B) Kit225 cells were made quiescent in
IL-2-free medium overnight and then stimulated with IL-2 or IL-9 for the indicated times. Phosphorylation of Jak3 at Y904 and Y939 was
monitored as described for panel A. Shown are representative data from two independent experiments. (C) Purified human T lymphocytes were
activated with PHA for 72 h, subsequently made quiescent for 24 h, and then stimulated with IL-2 for the indicated times. Phosphorylation of Jak3
on Y904 and Y939 was determined by Western blotting with the respective phospho-specific antibodies. Representative data from three
independent experiments are shown.
VOL. 28, 2008TYROSINES 904 AND 939 REGULATE Jak32275
determine whether Jak3 is phosphorylated at Y904 and Y939
in response to other ?c-containing cytokines, Kit225 cells were
stimulated with IL-9 as described in the legend to Fig. 3B. Jak3
protein was immunoprecipitated, and Western blotting was
performed with these two phospho-specific antibodies. We ob-
served that IL-9 was able to induce Jak3 phosphorylation at
Y904 and Y939, although the strength and kinetics of phos-
phorylation on these sites differed from the treatment with
IL-2. Nonetheless, these results suggest that ?c cytokines, in
general, can result in Jak3 phosphorylation at these sites.
To test whether Y904 and Y939 are phosphorylated in non-
tumorigenic primary lymphocytes, PHA-activated primary hu-
man T cells were made quiescent and then stimulated with
IL-2. IL-2-stimulated Jak3 phosphorylation of Y904 and Y939,
although less robust than that in tumor cell lines (panels A and
B), was detectable within 2 min before returning to basal levels
FIG. 4. Jak3 Y904F and Y939F variants have decreased autophosphorylation and kinase activity toward the exogenous substrate GST-?c.
(A) HEK293 cells were transfected with plasmids for WT Jak3 or the Y939F, Y939F, or K855A variant. Jak3 was immunoprecipitated (IP) from
cell lysates and tested for kinase activity in the presence of the Jak3 substrate GST-?c. The pY levels of Jak3 and GST-?c were assessed by
quantitative anti-pY antibody Western blotting (immunoblotting [IB]) normalized to Jak3 and GST-?c. (B) Phosphorylation of GST-?c was
quantified, and data from two representative independent experiments were plotted. (C) Jak3 autophosphorylation was quantified, and data from
two representative independent experiments were plotted.
2276CHENG ET AL.MOL. CELL. BIOL.
at 60 min (Fig. 3C). Moreover, IL-2-induced phosphorylation
of Y904 and Y939 had kinetics similar to those of total tyrosine
phosphorylation of Jak3, as detected by anti-pY blotting, ex-
cept that phosphorylation did not return to basal levels after 60
min. Thus, the phosphorylation of Y904 and Y939 occurred in
multiple cell types, including primary human T cells, with ac-
tivation profiles indicating a universal mechanism of Jak3
Y904 and Y939 are required for optimal Jak3 autophos-
phorylation and kinase activity in vitro. Phosphorylation rep-
resents an important posttranslational modification that regu-
lates the catalytic activities of protein kinases. To determine
whether phosphorylation at tyrosine 904 or 939 affects the
kinase activity of Jak3, HEK293 cells were transfected with
plasmids encoding WT Jak3 or the Y904F, Y939F, or K855A
mutant form, and these Jak3 proteins were subsequently im-
munoprecipitated and tested for autophosphorylation, as well
as kinase activity toward the exogenous substrate GST-?c. The
tyrosine phosphorylation of Jak3 and GST-?c was measured by
quantitative Western blotting with anti-pY antibody that was
normalized to the total amounts of Jak3 or GST-?c. Jak3 and
GST-?c levels were detected with anti-Jak3 and anti-GST an-
In these assays, WT Jak3 exhibited robust, time-dependent
autophosphorylation, as well as kinase activity on the GST-?c
substrate (Fig. 4A). This activity originated from immunopre-
cipitated Jak3, since no phosphorylation of GST-?c was ob-
served in immunoprecipitates of the Jak3 kinase-inactive
K855A mutant form. Importantly, in contrast to WT Jak3, both
the Y904F and Y939F mutant forms showed nearly 60% less
autophosphorylation (Fig. 4C) and kinase activity toward
GST-?c (Fig. 4B). These results suggest that both Y904 and
Y939 are vital for Jak3 autophosphorylation and the kinase
activity toward exogenous substrates. Hence, these decreased
kinase activities suggest that phosphorylation of Jak3 at Y904
or Y939 positively regulates this enzyme.
Jak3 Y904 and Y939 are required for optimal Stat5 tyrosine
phosphorylation and transcriptional activity. Stat5a and
Stat5b are two of seven Stat family members that share high
homology due to a gene duplication event (11). Human
Stat5a is tyrosine phosphorylated at Y694, while human Stat5b
is tyrosine phosphorylated at Y699, by Jak proteins and other
tyrosine kinases. Phosphorylation of cytokine receptor-based
tyrosines recruits Stat5, allowing their subsequent tyrosine
phosphorylation, which is required for their dimerization, nu-
clear translocation, and gene regulation activity (25). To de-
termine whether Y904 and Y939 of Jak3 are required for
optimal Stat5 activation in vivo, plasmids encoding WT Jak3 or
Y904F, Y939F, Y981F, or K855A mutant Jak3 were cotrans-
fected into HEK293 cells with an expression plasmid for
Stat5a. Stat5a was immunoprecipitated and tested for tyrosine
phosphorylation by Western blotting with an anti-pY Stat5
antibody that recognized pY694 in Stat5a (Fig. 5A). For these
assays, coexpression of WT Jak3 with Stat5a was found to
result in tyrosine phosphorylation of Stat5a (lane 2) mediated
by Jak3, since Stat5a coexpressed with the kinase-inactive
K855A mutant form of Jak3 was not tyrosine phosphorylated
(lane 6). As expected, expression of the hyperactive Y981F
mutant form of Jak3 increased Stat5a tyrosine phosphorylation
dramatically (lane 5). In contrast, tyrosine phosphorylation of
Stat5a was reduced nearly 50% for the Y904F mutant form
compared to WT Jak3 (lane 3) and was almost completely
abolished for the Y939F mutant form (lane 4). Reprobing this
blot with anti-Stat5a antibody confirmed that similar amounts
of Stat5a protein were tested and measured (Fig. 5A, lower
panel). As an additional control, Jak3 was also immunopre-
cipitated and tested for its levels of total tyrosine phosphory-
lation by blotting with anti-pY antibody (Fig. 5B). In contrast
FIG. 5. Jak3 Y939 and Y904 are required for optimal Stat5 activity
in vivo. (A) HEK293 cells were cotransfected with plasmids encoding
the Stat5a and Jak3 proteins as indicated. At 30 h posttransfection,
cells were harvested and cell lysates were immunoprecipitated (IP)
with anti-Stat5a antibody. Stat5a activation was then assessed by West-
ern blotting (immunoblotting [IB]) with anti-pY Stat5 antibody. Total
Stat5a levels were determined by reprobing the membrane with anti-
Stat5a antibody, as shown in the panel below. (B) Cell lysates were also
immunoprecipitated with antibodies to Jak3 and Western blotted for
tyrosine phosphorylation. Total Jak3 levels were monitored by reprob-
ing with anti-Jak3 antibodies as shown in the panel below. Shown are
representative data from four independent experiments. (C) In six-well
dishes, subconfluent cells were transfected in triplicate with the Stat5-
activated ?-casein–luciferase reporter plasmid, the TK-Renilla lucifer-
ase vector, Stat5a, and WT Jak3 or Jak3 mutant forms, as indicated.
Control (CTRL) wells were transfected with identical amounts of
luciferase and empty vectors. Cells were lysed, and the luciferase
activities were determined at 30 h posttransfection. Representative
data from three independent experiments are shown. Cell lysates were
also immunoblotted with anti-pY Stat5, anti-Stat5a, and anti-Jak3 an-
tibodies to verify equivalent expression of Jak3 and Stat5a (lower
VOL. 28, 2008TYROSINES 904 AND 939 REGULATE Jak3 2277
to WT Jak3 (lane 2), the Jak3 Y981F mutant form (lane 5)
showed more pronounced tyrosine phosphorylation, while the
tyrosine phosphorylation of Y904F mutant Jak3 (lane 3) was
modestly decreased and tyrosine phosphorylation of Y939F
mutant Jak3 (lane 4) was significantly reduced. Similar levels of
Jak3 protein were expressed for each variant, as determined by
Western blotting with anti-Jak3 antibodies (Fig. 5B, lower
panel). These data further support the model in which Y904
and Y939 are required for optimal Jak3 activation and subse-
quent tyrosine phosphorylation of endogenous substrates such
Tyrosine-phosphorylated Stat proteins dissociate from the
receptor complex to form dimers, translocate to the nucleus,
and cooperate with other factors to stimulate gene transcrip-
tion (24). To analyze the functional role of Jak3 Y904 and
Y939 in Stat5-dependent transcriptional activity, HEK293 cells
were transfected with Stat5a and the appropriate Jak3 expres-
sion plasmids, along with a ?-casein–firefly luciferase reporter
construct harboring three Stat5 response elements in tandem.
Cells were also transfected with a second Renilla luciferase
gene under the control of a constitutive thymidine kinase pro-
moter to control for transfection efficiency. In these experi-
ments, WT Jak3 promoted a sixfold induction of Stat5 acti-
vated ?-casein–luciferase reporter activity compared to the
Jak3 kinase-inactive K855A mutant form. In contrast to WT
Jak3, Y904F mutant Jak3 showed a 40% decrease in this Stat5
reporter activity, while a Y939F Jak3 variant was almost com-
pletely inactive (Fig. 5C). These data indicate that Jak3 Y904
and Y939 are required for activation of Stat5 transcriptional
activity. As a control, cell lysates were immunoblotted with
anti-pYStat5, anti-Stat5a, and anti-Jak3 antibodies to confirm
activation and expression (Fig. 5C, lower panel). These results
demonstrate that the reduced levels of Stat5a tyrosine phos-
phorylation, when cotransfected with the Y904F and Y939F
Jak3 variants, resulted in a direct reduction in its transcrip-
tional activity (lanes 2 to 4). Interestingly, the Jak3 Y981F
mutant form did not induce significantly higher luciferase gene
expression compared to WT Jak3, although the Jak3 Y981F
variant mediated substantially stronger tyrosine phosphoryla-
tion of Stat5a (lane 5).
Y904 and Y939 are localized to distinct regions of the Jak3
kinase domain and conserved among other Jak family mem-
bers. The Jak3 kinase domain has been crystallized in a com-
plex with the staurosporine analogue ANF941 (4). Based upon
this structure, Y904 and Y939 were localized in Jak3 to predict
their potential roles in Jak3 catalytic activity. Both tyrosine
residues are at least partially exposed to the solvent. Y904
localizes to the N lobe of the kinase domain. Interestingly, the
two residues flanking tyrosine 904, glutamate 903 and leucine
905, make direct contact with ANF941 (Fig. 6A). Since
ANF941 is also an ATP analogue, it seemed plausible to expect
that tyrosine 904 may be important for regulating ATP binding
during catalytic reactions. Conversely, Y939 is localized to
?-helix E in the C lobe of the kinase domain (Fig. 6B). Typi-
FIG. 6. Y904 and Y939 are located within distinct regions of the Jak3 catalytic domain and are conserved among Jak family members. (A) Y904
localizes to the ATP binding pocket of the N lobe of the kinase domain. (B) Y939 is harbored within ?-helix E of the C lobe of the kinase domain.
Models were made with the DeepView Swiss-pdbviewer 3.7 program, based on the crystal structure of the Jak3 kinase domain solved in complex
with the staurosporine analogue ANF941 (4) (accession number 1YVJ). Residues E903, Y904, L905, and Y939 and ANF941 are indicated in color
beneath the Jak3 structure. (C) The amino acid sequences surrounding the novel phosphotyrosine sites in human Jak3 were aligned with human
Jak1, Jak2, and Tyk2.
2278CHENG ET AL.MOL. CELL. BIOL.
cally, this kinase region promotes substrate access to the cat-
alytic cleft, suggesting that phosphorylation of Y939 may reg-
ulate the interaction between Jak3 and Stat5 or other
substrates. In addition, phosphorylation of Y939 creates a po-
tential Stat5 SH2 binding site (pY[VLTFIC]XX) that may
mediate high-affinity interaction with Stat5 (32).
Alignment of the amino acid sequences of the four human
Jak proteins revealed that Y904 is conserved in Jak2, Tyk2, and
Jak3 but is replaced by a phenylalanine in Jak1 (Fig. 6C). Since
this is a highly phosphorylated residue in Jak3, it is tempting to
speculate that this substitution may explain the lower level of
cytokine-stimulated tyrosine phosphorylation of Jak1 com-
pared to Jak3 which has been reported (22, 26). On the other
hand, Y939 is conserved among the four Jak proteins (Fig. 6C)
and also some other receptor tyrosine kinases, including the
epidermal growth factor receptor family members EGFR,
HER2, and ErbB4 (not shown). Taken together, this model
suggests that phosphorylation of this conserved tyrosine may
regulate Jak family kinases through a common mechanism.
Phenylalanine substitution of Y904 in Jak3 impairs its ATP
binding affinity. To determine whether Y904 in Jak3 is impor-
tant for its interaction with ATP, the ATP binding affinity of
WT Jak3 or the Y904F or Y939F mutant form of Jak3 was
examined. HEK293 cells were transfected with WT Jak3 or the
Y904F or Y939F mutant form, and the Jak3 proteins were
immunoprecipitated and tested for kinase activity toward
GST-?c in the presence of gradually increasing concentrations
of ATP. Tyrosine phosphorylation of GST-?c was assessed by
quantitative Western blotting with anti-pY antibodies (Fig.
7A). Blots were also probed with anti-GST and anti-Jak3 an-
tibodies to normalize protein levels. Data quantified from sev-
eral experiments are shown in Fig. 7B. Consistent with our
finding that Y904 and Y939 are required for optimal in vitro
catalytic activity of Jak3 (Fig. 4), the Vmaxfor the Y904F or
Y939F mutant form was decreased by 50% compared to that
for WT Jak3. In addition, the KmATPfor Jak3 was increased
from 1.31 ?M for the WT to 4.52 ?M for the Y904F variant,
while WT Jak3 and the Y939F mutant form showed similar
ATP binding affinities (the KmATPfor Y939F mutant Jak3 was
1.79 ?M). These data suggest that Y904 or Y939 can affect the
ability of Jak3 to consume ATP.
Stat5 binds to Y939 phosphorylated peptide. The previous
data (Fig. 5) support the notion that phosphorylation of Y939
in Jak3 is essential for Stat5 tyrosine phosphorylation and
transcriptional activity. However, in vitro data indicated that
the kinase activity of the Jak3 Y939F mutant form toward
GST-?c was only reduced approximately 60% compared to
that of WT Jak3 (Fig. 4). In addition to ATP binding, this
suggests that Y939 may be important for optimal interaction
between Jak3 and Stat5. To determine the possible direct reg-
ulatory mechanism of Jak3 Y939 in Stat5 activation, peptides
containing phosphorylated or nonphosphorylated Y939 were
coupled to CNBr-activated Sepharose and used in a pull-down
assay with YT-cell lysates. Captured Stat5 and pY Stat5 were
examined by Western blotting. Indeed, endogenous Stat5 was
found to bind the phosphorylated Y939 peptide but not its
nonphosphorylated counterpart (Fig. 8, upper panel). Simi-
larly, the phosphorylated Y904 peptide also captured Stat5
(data not shown). In either instance, this occurred whether or
not the cells had been stimulated with IL-2. Interestingly,
reprobing the membrane with anti-pY antibodies indicated
that additional tyrosine-phosphorylated proteins other than
Stat5 coprecipitated with the phosphorylated Y939 peptide in
an IL-2-dependent manner (Fig. 8, lower panel). These data
correlate with previous reports that peptides derived from the
corresponding phosphorylation site in Jak2, Y966, bound to a
number of signaling proteins containing SH2 domains, such as
Stat5, SHC, phosphatidylinositol 3-kinase, and a novel protein
tentatively named p70 (5). Taken as a whole, these data indi-
cate that, in addition to regulating the intrinsic catalytic activity
of Jak3, phosphorylation of Y939 and Y904 may provide a
docking site for interaction with Jak3 substrates such as Stat5.
FIG. 7. The Jak3 Y904F mutant form shows impaired ATP binding
activity. (A) HEK293 cells were transfected with plasmids for WT Jak3
or the Y904F or Y939F mutant form. Jak3 immunoprecipitation (IP)
kinase assays were performed with increasing concentrations of ATP
in the presence of 2 ?g GST-?c for 20 min as indicated. pY levels of
GST-?c were assessed by quantitative Western blotting (immunoblot-
ting [IB]) with anti-pY and anti-GST antibodies. Jak3 expression levels
were determined by Western blotting with anti-Jak3 antibody.
(B) Quantification of the KmATPand Vmaxfor WT Jak3 and the Y904F
and Y939F variants were performed with GraphPad Prism 4 software
and Michaelis-Menten kinetics. Each datum point represents at least
three independent experiments. Error bars represent the standard
error of the mean.
VOL. 28, 2008TYROSINES 904 AND 939 REGULATE Jak32279
Jak3 is a key immunoregulatory enzyme responsible for pro-
moting normal and abnormal immune responses. Thus, under-
standing its mechanism of regulation is fundamentally impor-
tant. For this work, two novel tyrosine phosphorylation sites in
Jak3 were identified by mass spectrometry and further charac-
terized with site-specific phospho-specific antibodies. Y904
and Y939 were determined to be rapidly and transiently phos-
phorylated in response to IL-2 in YT, Kit225, and primary
human T cells. IL-9 similarly activated the phosphorylation of
Jak3 at Y904 and Y939 in Kit225 cells. The tyrosine phospho-
rylation kinetics of Y904 or Y939 peaked at 2 min and returned
to baseline by 60 min in YT and human T lymphocytes. In
contrast, Kit225 cells showed protracted phosphorylation ki-
netics, with pY904 and pY939 only returning to basal levels
after 120 min of IL-2 stimulation. Additionally, Y904 and Y939
were readily autophosphorylated and required for Jak3 cata-
lytic activity to phosphorylate a defined substrate. This is likely
due to the requirements for Y904 and Y939 in ATP and sub-
For protein kinases, phosphorylation of key residues within
the activation loop, which is localized between subdomains VII
and VIII of the kinase domain, induces a conformational
change which facilitates the access of substrates to the active
site. Hence, phosphorylation of these residues often increases
the intrinsic catalytic activities of kinases (29), as is true for
Y1162 in the insulin receptor kinase (13). For Jak family pro-
teins, two adjacent tyrosine residues within the activation loop
have been implicated in the control of their catalytic activities.
In Jak3, Y980 is a positive regulatory site while Y981 nega-
tively controls its activity (47). Similarly, mutation of position-
ally conserved Y1007 to phenylalanine in Jak2 blocked its
activation while phenylalanine substitution of Y1008 had no
Jak proteins have seven regions of sequence similarity
named Janus homology (JH) domains. The tyrosine kinase
domain is localized to the C terminus in JH1. Jak proteins are
unique in having a catalytically inactive pseudokinase domain
(JH2), which has been shown to regulate their kinase activity
(36, 37). The N-terminal JH4 to JH7 domains of Jak proteins
are involved in receptor association. This region has a band 4.1,
ezrin, radixin, and moesin (FERM) domain (10). Several mu-
tations in the FERM domain of Jak3 were found in SCID
patients (9). Except for tyrosine phosphorylation within the
activation loop of the kinase domain, few studies have ex-
panded the functional roles of other putative phosphorylation
sites in Jak proteins. Among the four Jak kinases, autophos-
phorylation of Jak2 may be the best characterized. Indeed,
Y221 in the FERM domain and Y570 in the JH2 domain are
sites of autophosphorylation in Jak2 that differently regulate its
catalytic activity (3). Phosphorylation of Jak2 on Y119 in the
FERM domain was found to down-regulate erythropoietin sig-
naling by promoting the dissociation of Jak2 from the receptor
complex (8). Y813 in the JH2 domain of Jak2 represents an-
other site of phosphorylation which is required for binding to
the adaptor protein SH2-B?, which further enhances the ac-
tivity of this enzyme. Positionally conserved corresponding ty-
rosine 785 in Jak3 is phosphorylated in response to IL-2 and is
also important for binding to SH2-B? (23). Phosphorylation of
Jak2 on S523 was recently demonstrated to function as a reg-
ulatory feedback mechanism to dampen the activity of this
enzyme (28). While little is known about serine phosphoryla-
tion in other Jak proteins, we also observed that Jak3 was
serine phosphorylated upon IL-2 stimulation (data not shown).
Additional work seeks to identify this serine residue(s).
Extending the current model, our data indicate that phos-
phorylation of both Y904 and Y939 positively regulates Jak3
activity. This is based on our observations that mutation of
either site to phenylalanine impairs the autocatalytic activity of
Jak3, as well as its ability to phosphorylate substrates such as
GST-?c in vitro (Fig. 4). Phenylalanine mutation of either
Y904 or Y939 impaired its ability to stimulate Stat5 tyrosine
phosphorylation and transcriptional activity (Fig. 5). Our data
provide insight into the molecular mechanisms by which phos-
phorylation of Y904 and Y939 positively regulates Jak3 activ-
ity. Based upon the crystal structure of the Jak3 kinase domain,
Y904 is likely confined to the ATP binding pocket between two
amino acids shown to make contact with the ATP analogue
ANF941(Fig. 6). Indeed, phenylalanine substitution of Y904
increased the Kmof Jak3 for ATP from 1.31 to 4.52 ?M and
reduced the Vmaxof Jak3 toward GST-?c by approximately
50% (Fig. 7). Although Y939 lies outside the proposed ATP
binding domain, this residue also changed the Kmof Jak3 for
ATP from 1.31 to 1.79 ?M and reduced the Vmaxof Jak3
toward GST-?c by approximately 50%. Nonetheless, these
data indicate that Y904 and Y939 are important for Jak3 ATP
FIG. 8. Association of Stat5 with phosphorylated Y939 peptides.
Peptides containing nonphosphorylated or phosphorylated Y939 were
coupled to CNBr-activated beads and used to probe for interacting
proteins from YT-cell lysates. YT cells (3 ? 107/sample) were left
untreated or stimulated with 100 nM IL-2 for 10 min. Soluble cell
lysates were incubated with the indicated peptides for 1 h at 4°C. After
washing, bound proteins were eluted from the beads with SDS sample
buffer. Binding of Stat5, pY Stat5, and other tyrosine-phosphorylated
proteins was assessed by Western blotting (immunoblotting [IB]) with
anti-Stat5, anti-pY Stat5, and anti-pY antibodies. Molecular mass
markers are indicated on the left. One representative set of data from
three independent experiments is shown.
2280CHENG ET AL.MOL. CELL. BIOL.
binding affinity and that phosphorylation of these sites pro-
motes Jak3 catalytic activity.
Interestingly, Y939 is localized to the ?-helix E of the C lobe,
suggesting that phosphorylation of this site may influence sub-
strate interaction. Moreover, phosphorylation of this residue is
predicted to create a Stat5 SH2 binding site (pY[VLTFIC]XX).
To test this notion, phenylalanine substitution of Y939 in Jak3
was found to result in reduced catalytic activity in vitro (Fig. 4)
and completely blocked its ability to activate Stat5a in vivo (Fig.
5). Moreover, a 10-mer peptide harboring pY939, but not its
nonphosphorylated counterpart, was able to capture endogenous
Stat5 from YT-cell lysates (Fig. 8). It is also noteworthy that this
peptide readily coassociated with other, unknown, tyrosine-phos-
phorylated proteins, indicating that phosphorylated Y939 may
regulate other signaling molecules. These data lead us to propose
that Y939 serves two roles in the Jak3 activation mechanism.
When phosphorylated, it may enhance access to the catalytic cleft
provide a docking site for SH2 domain-containing proteins such
as Stat5. Results in Fig. 5 indicate that Y904 is less effective than
Y939 in mediating Stat5 activation. These data may explain how
involvement (23, 47). Based on our findings, we hypothesize that,
in the absence of cytokine receptors, phosphorylation of Jak3 at
Y939, and possibly Y904, is required for Stat5 association and
subsequent activation. This may be especially important in tumor
models where constitutively active Jak3 may activate substrates
independently of a receptor.
In summary, we have identified Y904 and Y939 in Jak3 as
two novel sites of cytokine-mediated phosphorylation. Phos-
phoantibodies that specifically recognized these residues con-
firmed that Jak3 Y904 and Y939 are rapidly and transiently
induced in YT, Kit225, and primary human T cells. Y904 and
Y939 positively regulate the enzymatic activity of Jak3 and its
substrate Stat5. Lastly, we provide evidence that phosphoryla-
tion of Y904 and Y939 of Jak3 regulates Stat5 and ATP bind-
ing activity. These data provide new insight into the mecha-
nisms that regulate Jak3 activation and its downstream
This work was supported by grants from the Lizanell and Colbert
Coldwell Foundation; JP Morgan Chase Bank, N.A., Trustee; the
National Institutes of Health (AI053566); and the National Center for
Research Resources (5G12RR008124), a component of the National
Institutes of Health, to R. A. Kirken.
1. Aaronson, D. S., and C. M. Horvath. 2002. A road map for those who don’t
know JAK-STAT. Science 296:1653–1655.
2. Alexander, W. S., and D. J. Hilton. 2004. The role of suppressors of cytokine
signaling (SOCS) proteins in regulation of the immune response. Annu. Rev.
3. Argetsinger, L. S., J. L. Kouadio, H. Steen, A. Stensballe, O. N. Jensen, and
C. Carter-Su. 2004. Autophosphorylation of JAK2 on tyrosines 221 and 570
regulates its activity. Mol. Cell. Biol. 24:4955–4967.
4. Boggon, T. J., Y. Li, P. W. Manley, and M. J. Eck. 2005. Crystal structure of
the Jak3 kinase domain in complex with a staurosporine analog. Blood
5. Carpino, N., R. Kobayashi, H. Zang, Y. Takahashi, S. T. Jou, J. Feng, H.
Nakajima, and J. N. Ihle. 2002. Identification, cDNA cloning, and targeted
deletion of p70, a novel, ubiquitously expressed SH3 domain-containing
protein. Mol. Cell. Biol. 22:7491–7500.
6. Changelian, P. S., M. E. Flanagan, D. J. Ball, C. R. Kent, K. S. Magnuson,
W. H. Martin, B. J. Rizzuti, P. S. Sawyer, B. D. Perry, W. H. Brissette, S. P.
McCurdy, E. M. Kudlacz, M. J. Conklyn, E. A. Elliott, E. R. Koslov, M. B.
Fisher, T. J. Strelevitz, K. Yoon, D. A. Whipple, J. Sun, M. J. Munchhof, J. L.
Doty, J. M. Casavant, T. A. Blumenkopf, M. Hines, M. F. Brown, B. M. Lillie,
C. Subramanyam, C. Shang-Poa, A. J. Milici, G. E. Beckius, J. D. Moyer, C.
Su, T. G. Woodworth, A. S. Gaweco, C. R. Beals, B. H. Littman, D. A. Fisher,
J. F. Smith, P. Zagouras, H. A. Magna, M. J. Saltarelli, K. S. Johnson, L. F.
Nelms, S. G. Des Etages, L. S. Hayes, T. T. Kawabata, D. Finco-Kent, D. L.
Baker, M. Larson, M. S. Si, R. Paniagua, J. Higgins, B. Holm, B. Reitz, Y. J.
Zhou, R. E. Morris, J. J. O’Shea, and D. C. Borie. 2003. Prevention of organ
allograft rejection by a specific Janus kinase 3 inhibitor. Science 302:875–878.
7. Feng, J., B. A. Witthuhn, T. Matsuda, F. Kohlhuber, I. M. Kerr, and J. N.
Ihle. 1997. Activation of Jak2 catalytic activity requires phosphorylation of
Y1007 in the kinase activation loop. Mol. Cell. Biol. 17:2497–2501.
8. Funakoshi-Tago, M., S. Pelletier, T. Matsuda, E. Parganas, and J. N. Ihle.
2006. Receptor specific downregulation of cytokine signaling by autophos-
phorylation in the FERM domain of Jak2. EMBO J. 25:4763–4772.
9. Gadina, M., D. Hilton, J. A. Johnston, A. Morinobu, A. Lighvani, Y. J. Zhou,
R. Visconti, and J. J. O’Shea. 2001. Signaling by type I and II cytokine
receptors: ten years after. Curr. Opin. Immunol. 13:363–373.
10. Girault, J. A., G. Labesse, J. P. Mornon, and I. Callebaut. 1999. The N-
termini of FAK and JAKs contain divergent band 4.1 domains. Trends
Biochem. Sci. 24:54–57.
11. Grimley, P. M., F. Dong, and H. Rui. 1999. Stat5a and Stat5b: fraternal twins
of signal transduction and transcriptional activation. Cytokine Growth Fac-
tor Rev. 10:131–157.
12. Hou, S. X., Z. Zheng, X. Chen, and N. Perrimon. 2002. The Jak/STAT
pathway in model organisms: emerging roles in cell movement. Dev. Cell
13. Hubbard, S. R., L. Wei, L. Ellis, and W. A. Hendrickson. 1994. Crystal
structure of the tyrosine kinase domain of the human insulin receptor.
14. Ihle, J. N. 2001. The Stat family in cytokine signaling. Curr. Opin. Cell Biol.
15. Ihle, J. N., W. Thierfelder, S. Teglund, D. Stravopodis, D. Wang, J. Feng, and
E. Parganas. 1998. Signaling by the cytokine receptor superfamily. Ann.
N. Y. Acad. Sci. 865:1–9.
16. Ihle, J. N., B. A. Witthuhn, F. W. Quelle, K. Yamamoto, and O. Silvennoinen.
1995. Signaling through the hematopoietic cytokine receptors. Annu. Rev.
17. Irie-Sasaki, J., T. Sasaki, W. Matsumoto, A. Opavsky, M. Cheng, G. Welstead,
E. Griffiths, C. Krawczyk, C. D. Richardson, K. Aitken, N. Iscove, G. Koretzky,
P. Johnson, P. Liu, D. M. Rothstein, and J. M. Penninger. 2001. CD45 is a JAK
phosphatase and negatively regulates cytokine receptor signalling. Nature 409:
18. Johnston, J. A., C. M. Bacon, M. C. Riedy, and J. J. O’Shea. 1996. Signaling
by IL-2 and related cytokines: JAKs, STATs, and relationship to immuno-
deficiency. J. Leukoc. Biol. 60:441–452.
19. Johnston, J. A., M. Kawamura, R. A. Kirken, Y. Q. Chen, T. B. Blake, K.
Shibuya, J. R. Ortaldo, D. W. McVicar, and J. J. O’Shea. 1994. Phosphor-
ylation and activation of the Jak-3 Janus kinase in response to interleukin-2.
20. Kawamura, M., D. W. McVicar, J. A. Johnston, T. B. Blake, Y.-Q. Chen,
B. K. Lal, A. R. Lloyd, D. J. Kelvin, J. E. Staples, J. R. Ortaldo, and J. J.
O’Shea. 1994. Molecular cloning of L-JAK, a Janus family protein-tyrosine
kinase expressed in natural killer cells and activated leukocytes. Proc. Natl.
Acad. Sci. USA 91:6374–6378.
21. Kirken, R. A., R. A. Erwin, D. Taub, W. J. Murphy, F. Behbod, L. Wang, F.
Pericle, and W. L. Farrar. 1999. Tyrphostin AG-490 inhibits cytokine-medi-
ated JAK3/STAT5a/b signal transduction and cellular proliferation of anti-
gen-activated human T cells. J. Leukoc Biol. 65:891–899.
22. Kirken, R. A., H. Rui, M. G. Malabarba, O. M. Howard, M. Kawamura, J. J.
O’Shea, and W. L. Farrar. 1995. Activation of JAK3, but not JAK1, is critical
for IL-2-induced proliferation and STAT5 recruitment by a COOH-terminal
region of the IL-2 receptor beta-chain. Cytokine 7:689–700.
23. Kurzer, J. H., L. S. Argetsinger, Y.-J. Zhou, J. L. Kouadio, J. J. O’Shea, and
C. Carter-Su. 2004. Tyrosine 813 is a site of JAK2 autophosphorylation
critical for activation of JAK2 by SH2-B?. Mol. Cell. Biol. 24:4557–4570.
24. Levy, D. E., and J. E. Darnell, Jr. 2002. Stats: transcriptional control and
biological impact. Nat. Rev. Mol. Cell. Biol. 3:651–662.
25. Lin, J. X., and W. J. Leonard. 2000. The role of Stat5a and Stat5b in signaling
by IL-2 family cytokines. Oncogene 19:2566–2576.
26. Malabarba, M. G., R. A. Kirken, H. Rui, K. Koettnitz, M. Kawamura, J. J.
O’Shea, F. S. Kalthoff, and W. L. Farrar. 1995. Activation of JAK3, but not
JAK1, is critical to interleukin-4 (IL4) stimulated proliferation and requires
a membrane-proximal region of IL4 receptor ?. J. Biol. Chem. 270:9630–
27. Malabarba, M. G., H. Rui, H. H. Deutsch, J. Chung, F. S. Kalthoff, W. L.
Farrar, and R. A. Kirken. 1996. Interleukin-13 is a potent activator of JAK3
and STAT6 in cells expressing interleukin-2 receptor-? and interleukin-4
receptor-?. Biochem. J. 319(Pt. 3):865–872.
28. Mazurkiewicz-Munoz, A. M., L. S. Argetsinger, J. L. Kouadio, A. Stensballe,
O. N. Jensen, J. M. Cline, and C. Carter-Su. 2006. Phosphorylation of JAK2
VOL. 28, 2008TYROSINES 904 AND 939 REGULATE Jak3 2281
at serine 523: a negative regulator of JAK2 that is stimulated by growth
hormone and epidermal growth factor. Mol. Cell. Biol. 26:4052–4062.
29. Morgan, D. O., and H. L. De Bondt. 1994. Protein kinase regulation: insights
from crystal structure analysis. Curr. Opin. Cell Biol. 6:239–246.
30. Musso, T., J. A. Johnston, D. Linnekin, L. Varesio, T. K. Rowe, J. J. O’Shea,
and D. W. McVicar. 1995. Regulation of JAK3 expression in human mono-
cytes: phosphorylation in response to interleukins 2, 4, and 7. J. Exp. Med.
31. Pernis, A. B., and P. B. Rothman. 2002. JAK-STAT signaling in asthma.
J. Clin. Investig. 109:1279–1283.
32. Puntervoll, P., R. Linding, C. Gemund, S. Chabanis-Davidson, M. Mattingsdal,
S. Cameron, D. M. Martin, G. Ausiello, B. Brannetti, A. Costantini, F. Ferre, V.
Maselli, A. Via, G. Cesareni, F. Diella, G. Superti-Furga, L. Wyrwicz, C. Ramu,
C. McGuigan, R. Gudavalli, I. Letunic, P. Bork, L. Rychlewski, B. Kuster, M.
Helmer-Citterich, W. N. Hunter, R. Aasland, and T. J. Gibson. 2003. ELM
server: a new resource for investigating short functional sites in modular eu-
karyotic proteins. Nucleic Acids Res. 31:3625–3630.
33. Ross, J. A., Z. S. Nagy, H. Cheng, S. M. Stepkowski, and R. A. Kirken. 2007.
Regulation of T cell homeostasis by JAKs and STATs. Arch. Immunol. Ther.
34. Russell, S. M., J. A. Johnston, M. Noguchi, M. Kawamura, C. M. Bacon, M.
Friedmann, M. Berg, D. W. McVicar, B. A. Witthuhn, O. Silvennoinen, et al.
1994. Interaction of IL-2R beta and gamma c chains with Jak1 and Jak3:
implications for XSCID and XCID. Science 266:1042–1045.
35. Russell, S. M., N. Tayebi, H. Nakajima, M. C. Riedy, J. L. Roberts, M. J.
Aman, T. S. Migone, M. Noguchi, M. L. Markert, R. H. Buckley, J. J.
O’Shea, and W. J. Leonard. 1995. Mutation of Jak3 in a patient with SCID:
essential role of Jak3 in lymphoid development. Science 270:797–800.
36. Saharinen, P., and O. Silvennoinen. 2002. The pseudokinase domain is
required for suppression of basal activity of Jak2 and Jak3 tyrosine kinases
and for cytokine-inducible activation of signal transduction. J. Biol. Chem.
37. Saharinen, P., M. Vihinen, and O. Silvennoinen. 2003. Autoinhibition of
Jak2 tyrosine kinase is dependent on specific regions in its pseudokinase
domain. Mol. Biol. Cell 14:1448–1459.
38. Simoncic, P. D., A. Lee-Loy, D. L. Barber, M. L. Tremblay, and C. J.
McGlade. 2002. The T cell protein tyrosine phosphatase is a negative regu-
lator of Janus family kinases 1 and 3. Curr. Biol. 12:446–453.
39. Stepkowski, S. M., J. Kao, M. E. Wang, N. Tejpal, H. Podder, L. Furian, J.
Dimmock, A. Jha, U. Das, B. D. Kahan, and R. A. Kirken. 2005. The
Mannich base NC1153 promotes long-term allograft survival and spares the
recipient from multiple toxicities. J. Immunol. 175:4236–4246.
40. Sugamura, K., H. Asao, M. Kondo, N. Tanaka, N. Ishii, K. Ohbo, M.
Nakamura, and T. Takeshita. 1996. The interleukin-2 receptor ? chain: its
role in the multiple cytokine receptor complexes and T cell development in
XSCID. Annu. Rev. Immunol. 14:179–205.
41. Tortolani, P. J., B. K. Lal, A. Riva, J. A. Johnston, Y. Q. Chen, G. H.
Reaman, M. Beckwith, D. Longo, J. R. Ortaldo, K. Bhatia, I. McGrath, J.
Kehrl, J. Tuscano, D. W. McVicar, and J. J. O’Shea. 1995. Regulation of
JAK3 expression and activation in human B cells and B cell malignancies.
J. Immunol. 155:5220–5226.
42. Wakao, H., F. Gouilleux, and B. Groner. 1994. Mammary gland factor
(MGF) is a novel member of the cytokine regulated transcription factor gene
family and confers the prolactin response. EMBO J. 13:2182–2191.
43. Wang, Y., M. G. Malabarba, Z. S. Nagy, and R. A. Kirken. 2004. Interleukin
4 regulates phosphorylation of serine 756 in the transactivation domain of
Stat6. Roles for multiple phosphorylation sites and Stat6 function. J. Biol.
44. Ward, A. C., I. Touw, and A. Yoshimura. 2000. The Jak-Stat pathway in
normal and perturbed hematopoiesis. Blood 95:19–29.
45. Witthuhn, B. A., O. Silvennoinen, O. Miura, K. S. Lai, C. Cwik, E. T. Liu,
and J. N. Ihle. 1994. Involvement of the Jak-3 Janus kinase in signalling by
interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370:153–157.
46. Yamashita, H., J. Xu, R. A. Erwin, W. L. Farrar, R. A. Kirken, and H. Rui.
1998. Differential control of the phosphorylation state of proline-juxtaposed
serine residues Ser725 of Stat5a and Ser730 of Stat5b in prolactin-sensitive
cells. J. Biol. Chem. 273:30218–30224.
47. Zhou, Y. J., E. P. Hanson, Y. Q. Chen, K. Magnuson, M. Chen, P. G. Swann,
R. L. Wange, P. S. Changelian, and J. J. O’Shea. 1997. Distinct tyrosine
phosphorylation sites in JAK3 kinase domain positively and negatively reg-
ulate its enzymatic activity. Proc. Natl. Acad. Sci. USA 94:13850–13855.
2282CHENG ET AL.MOL. CELL. BIOL.